Winspear Retractable LED Chandelier

TECHNICAL FOCUS: LIGHTING / RIGGING
Copyright Lighting&Sound America February 2010 http://www.lightingandsoundamerica.com/LSA.html
The Retractable LED
Chandelier
By: Mike Wood
It’s only a chandelier, isn’t it? Just a
few lights. What could be so technologically interesting about that? Well,
that’s what I was thinking almost two
years ago, when I first heard about
the proposed chandelier at the
Winspear Opera House in Dallas. (See
page 46 for the feature story.)
However, once I grasped the scale of
the proposal, it became clear that this
wasn’t your ordinary, everyday chandelier, if there is such a thing. It all
started with an artist’s concept picture
from the project’s architect, Foster +
Partners, and the architectural lighting
consultant, Claude R. Engle.
This showed a 40' diameter ethereal construction with seemingly unsupported bars of light hanging in a
roughly pyramidical shape from the
main auditorium ceiling. Discussion
revealed that the intent was to have
over 300 illuminated bars, which would
reach down 50' in front of the two top
balcony tiers in a three-dimensional
structure to provide a centerpiece for
the room while the audience was
being seated before the performance.
At show time, these bars would rise
until the ends were flush with the ceiling, where they would form a star field
before finally extinguishing at curtain
up. The bars themselves needed to be
illuminated as evenly as possible,
should match the warm white of the
existing incandescent house lighting,
and complement the white gold leaf
on the balcony fronts. The chandelier
was a key part of the room’s artistic
design and would be the center of
attention—both literally and metaphorically—of the audience before the curtain went up. Okay, that was the concept, but how do you actually build it?
Figure 1: Concept. (Photograph courtesy of Foster + Partners)
66 • February 2010 • Lighting&Sound America
The architects had considered a fiberoptic solution but had not been able to
find a way to raise and lower it 50'
while keeping the suspension invisible
to the audience and solely illuminating
the final 6' portion of each fiber.
J.R. Clancy was already heavily
involved in the Winspear installation.
As there was a significant rigging
implication to making this chandelier
raise and lower safely and quietly over
the heads of an audience, the company was also invited to bid for the chandelier addition, which left the project
manager, Robert Degenkolb, the task
of coordinating an engineering concept that would meet both the artistic
and engineering requirements. The
chandelier is primarily an artistic installation, with the illumination for the
room coming from conventional luminaires, so it needed an engineering
design that wouldn’t interfere with the
aesthetic concept and was essentially
invisible without compromising safety
and reliability. At that stage, J.R.
Clancy brought me in to assist with
development of the illuminated rods; it
turned into a joint project, with Clancy
taking the lead and providing all of the
hoists, control, structural support, rigging, and suspension while Mike
Wood Consulting and Scott Ingham, of
Ingham Designs, provided the illuminated rods, drive electronics, data distribution, programming, and control.
The first task for the team was to
develop the rods themselves, as
those would be the only portion the
audience would see, and, once they
were working, everything else in the
engineering design would follow from
that decision. The team experimented
All photographs by Mike Wood except where indicated
Engineering a centerpiece attraction at
the Winspear Opera House
with various materials and illumination
sources before ending up with an
acrylic rod and a custom four-channel
RGBW LED illuminator that was optically coupled to the top of the rod.
We tried many different types of surface finishes to ensure even illumination but, in the end, simplicity was
best, and a polished cast acrylic rod
gave the solid bar of light that was
needed. Cast acrylic performs much
better than an extruded version for
light transmission in this kind of application; the cast material is easier to
machine and form and doesn’t have
the entrained air bubbles and imperfections present in the die extruded
product. Those imperfections glow
like little stars within the rod; they’re
very pretty, but not what was wanted
here. A custom 6'-long cast rod, with
the ends polished optically flat, was
selected. For illumination, four standard 3W RGBW LED dies were packaged with a custom lens so that the
exit angle matched as closely as possible the TIR (total internal reflection)
angle of the acrylic; therefore, as little
light as possible was wasted in coupling to the top end. The use of TIR in
a solid rod is optically identical to the
fiber-optic solution that the lighting
consultant first envisaged; it’s just
that it is 318 separate 6'-long rigid
fibers rather than flexible bundles.
Although the main desire of the lighting consultant and architects was for
white light, as previously mentioned,
Figure 2: Rod testing.
the white had to be controllable to
get the right color temperature and
warmth to match the color scheme of
the room, so an RGBW array gave
the flexibility needed.
Figure 2 shows an early trial with
various patterns and surface finishes.
The plain polished rod (third from the
right) performed the best and met the
architect’s and lighting consultant’s
requirements.
Once a basic rod shape and style
had been agreed upon, the project
could proceed to a mock-up of the
Figure 3: Early trial using white ropes as a
proof of concept and scale.
(Photograph: Robert Degenkolb)
entire chandelier—or at least half of it!
J.R. Clancy took over a local theatre in
Syracuse for a day and constructed a
full-scale model using black cord and
white rope of the same diameter as
the acrylic rods. Some effective use of
the stage lighting installation completed the job, as shown in Figure 3.
With this 1:1 model to approve the
overall scale and, in particular, the
diameter and length of the illuminated
rods, the architects and owners
signed off on the project, and we had
the green light to proceed to a
detailed design.
Although we were using 3W LEDs,
we wanted to run them at very conservative levels—the installation needs to
last at least 10 years with daily usage,
and the key to getting a good life from
LEDs is good thermal management.
Each acrylic rod was capped with a
head, which formed the enclosure for
the LEDs and their associated drivers,
and also provided secure mechanical
suspension for the rod. The head was
also the heat sink for the LEDs; the
thermal design, using heat flow analysis, was specified to allow at least
50% headroom on the expected
power usage. The final head design
was an interesting collaboration
between the mechanical engineers,
whose main concern was the safety
aspects of hanging a 6' rod over the
audience; the electronics engineers,
with their requirements for heat flow,
EMI shielding and connectivity; and
the understanding by all that this was
in view of the audience and therefore
had to fit the aesthetics of the design.
It seemed like the simplest component
in the installation—just an aluminum
extrusion—but it turned out to be one
of the most complicated and critical
parts of the design.
Nothing on this project was simple—not even the cabling. Each rod
head was suspended from a single
custom cable, which contained a
shielded twisted pair for data, a pair
of wires for 24V DC, and three synthetic fiber suspension members. The
whole assembly was enclosed in a
matte black sheath to produce a final
cable that was just 0.25'' in diameter.
This cable had to pass over sheaves
and blocks onto a motor drum hoist
without breaking, so ultra-flex insulated conductor cables of a type more
commonly seen in moving lights were
used for the construction. Before
committing to the production run,
Clancy ran both proof load testing
and an accelerated simulated life test,
confirming that we could successfully
pass data and power through the
system over the life of the cable with
the specified connected load. The
head and rod only weighs about 5lbs;
however, we had a concern that there
could be problems if the rods ever
knocked and entangled with each
other, so minimum 8:1 safety factors
were used in the development of the
suspension system.
Now to the hoists: There are 318
rods in total, and the ideal situation
TECHNICAL FOCUS: LIGHTING / RIGGING
would be if there was one hoist per
rod. However, cost put that possibility
out of the running very quickly, and a
solution using fewer hoists had to be
found. In the final installation, the 318
rods are distributed over 44 lifting
mechanisms, each of which controls
up to eight rods. In the chandelier’s
Figure 4: Hoist and cables grid layout. (Drawing courtesy of J.R. Clancy)
Figure 5: A (not so) tangled web.
68 • February 2010 • Lighting&Sound America
operational shape, those 318 rods are
at widely different height bands, but
each hoist has to be connected to
eight rods, which are all at the same
position. To make this trickier, the
overall conical shape of the chandelier
means that rods at the same height
are never adjacent and the ceiling itself
isn’t flat; instead, it is both domed and
at an angle. To accommodate these
requirements, the hoist cable routing in
the grid from each hoist is incredibly
complex. A three-dimensional layout
had to be prepared (Fig. 4) so that
none of the cables interfered with each
other as they passed over or under
multiple other cables on their route
from hoist drum to loft block.
Figure 5 shows how that translated
to reality in the installation. After a lot
of work by skilled installers, it ended
up being remarkably tidy.
Also visible in Figure 5 are the tops
of tubes in the grid that provide guides
and protection for the rods when they
are parked in the void between the
plaster ceiling of the auditorium and
the grid. Figure 6 shows those tubes
as seen from the surrounding catwalks.
The tubes also contain twin stacks of
brushes, which serve to both keep the
rods clean and to provide a light block.
As mentioned above, each hoist
raises and lowers up to eight rods, and
every LED in each of those eight rods
can be individually controlled from data
sent down the cable. We had concerns
about hanging 318 long antennas in an
opera house and so wanted to minimize that data; we also wanted every
rod and its circuitry to be completely
identical and interchangeable. The best
solution was to have an intelligent data
distributor on every hoist, which took in
DMX512 data, then formatted it and
rebroadcast it down the eight cables,
with each rod only receiving its own
data. This has the added benefit that
the heads are all being fed with home
runs, so no addressing or networking is
needed. The net result is a vast reduction in the amount of data passing
down the cables, which mitigates the
risk of electromagnetic noise causing
interference with the sound systems.
That, coupled with the silent operation
of the hoists, meant that we passed
the scrutiny of the audio consultants—
pretty important in an opera house! To
further reduce data flow, each rod has
a processor and does all its own crossfading, interpolating, and smoothing of
the incoming eight-bit data to an internal 16-bit and adding pseudo thermal
delay so that the output matches an
incandescent dimming curve. Key to
maintaining the incandescent illusion is
avoiding the final “blink” as LEDs turn
off, so a great deal of care was taken
with the final 10% of dimming. This is
particularly important with house lights
as, by their nature, they nearly always
turn off into a blackout where the
slightest imperfection is apparent.
The data distributor is mounted on
the rotating portion of the hoist, with
power and data fed to it through a
data-quality commutator. The software
in the data distributor and heads
watch for any glitches in the data feed
through the commutator and hide it so
the output remains smooth. We were
helped in this by the application—with
normal entertainment lighting, speed
of response is critical but, in this case,
immediate response can take second
place to smooth operation. The crossfade speed is provided as a programmable operational parameter so, if the
Winspear staff wants to change the
fade rate through the ETC Mosaic system, they can.
Figure 7 shows some of the
installed J.R. Clancy hoists suspended
from a Unistrut grid that is attached to
the underside of the building structure.
At the front of each hoist is a mounting
frame diverter assembly that directs
each lift line to the ceiling tubes. The
diverters incorporate a spring-loaded
overload assembly that stops the individual hoist in the event a lift line
becomes snagged during operation.
The hoists are controlled by a JR
Clancy SceneControl 500 through a
wired touch-screen pendant interface.
The pendant plug-in stations are located so that the operator has a good
view of the moving rods at all times
and allows for individual, group, and
preset control of the rod positions.
Interestingly, the lighting control room
Figure 6: Guide tubes between grid and
plaster ceiling.
Figure 7: Hoists.
Figure 8: Finished chandelier.
is not normally one of those stations,
as it has no view of the chandelier!
The pendant also provides a wired Estop connection to the system and
allows for password-protected adjustment of the variable speed hoist drives
to reconfigure the system.
Control of the 1,590 channels driving the 318 illuminator heads (each
uses five control channels—four colors
and a timing channel) is provided
through two linked ETC Mosaic show
control systems which, in turn, are
triggered from the ETC Unison
Paradigm and AMX touch panels distributed throughout the building.
Programming made full use of
Mosaic’s ability to map an array of
channels to a matrix which dramatically simplified the process. Programming
each of those 1,590 channels by hand
would have been painful. Instead, we
were able to overlay moving patterns
and cloud imagery to give a constantly
shifting dynamic to the chandelier.
Figure 8 shows the final chandelier.
You can’t tell from the photo, of
course, but the programming is quite
subtle; each rod is continuously shifting its color and brightness on either
side of the defined warm white by a
small amount, and no channel is ever
stationary. This movement is not
enough to be overt, but it gives the
whole thing the kind of life that you get
with a candle flame chandelier. We
also programmed in red-shift as the
chandelier dims so that, again, it
matches the color shift in the incandescent house lights on the balcony
fronts. These fades are timed to match
the one-minute movement of the
chandelier as it flies out so that, when
the rods reach their top preset with
their bottom ends level with the ceiling, the rods have dimmed down to a
TECHNICAL FOCUS: LIGHTING / RIGGING
That’s as it should be—we are in the
business of theatricality and illusion,
after all.
The installation was supervised by
Robert J. Degenkolb, of JR Clancy,
and Mike Wood and Scott Ingham, of
Mike Wood Consulting and Ingham
Designs, LLC, respectively, and was
performed by Joel Svoboda, Ken
Eggers, Charlie Hulme, Byron Payne,
and the International Association of
Iron Workers Local 263 (Dallas/
Fort Worth).
Mike Wood provides technical and
intellectual property consulting services to the entertainment technology
industry. He can be reached at
[email protected].
Figure 9 - Star field.
very warm star field glimmer with a
superimposed twinkle (Fig. 9).
The Winspear Opera House is now
open, and the chandelier runs every
night in front of an audience with no
idea of the engineering complexities
needed to perform such a simple
effect. Instead, it’s an art installation
that, I hope, compares well with the
original artist’s concept of Figure 1.